LoomR tutorial doesn't seem to work.

[alexandruioanvoda]

Hi! Hope this finds you well!

Is the LoomR tutorial not up-to-date anymore? Had a bug right in the tutorial: https://satijalab.org/loomR/loomR_tutorial.html

> library(loomR)
Loading required package: R6
Loading required package: hdf5r
> library(Seurat)
> hca <- connect(filename = "cc95ff89-2e68-4a08-a234-480eca21ce79.homo_sapiens.loom", mode = "r+")
> NormalizeData(object = hca, chunk.size = 1000, scale.factor = 10000, display.progress = TRUE)
Warning: The following arguments are not used: chunk.size, display.progress
Suggested parameter: verbose instead of display.progress

Error in as(object = data, Class = "dgCMatrix") :
   no method or default for coercing “loom” to “dgCMatrix”

I'm essentially just trying to process/analyze the Human Cell Atlas Immune Census data (the file is here: https://data.humancellatlas.org/project-assets/project-matrices/cc95ff89-2e68-4a08-a234-480eca21ce79.homo_sapiens.loom), so if you've got any alternative tutorial for that, I'd welcome any help!

[MzwaneleN]

@alexandruioanvoda please let me know if you managed to work your way around this

[alexandruioanvoda]

@MzwaneleN I have not, is there any new/updated tutorial?

[MzwaneleN]

@alexandruioanvoda unfortunately it's still the same tutorial

[danaschwalbe]

Hi! Before trying the NormalizeData() function, you should try the following:

hca<-as.Seurat(hca,cells="CellID",features="Gene")

this will make the loom file a seurat object. The connect() function simply loads the loom file into R. Hope this helps!